Francisco J. Rodríguez-Cortés , Juan Antonio Arias-López , Manuel Oviedo-de la Fuente , Jose M. Jiménez-Pastor , Luna López-Coleto , Pedro Arévalo-Buitrago , Juan de la Cruz López-Carrasco , Rocío Valverde-León , Pablo Jesús López-Soto , Ignacio Morales-Cané
{"title":"Impact of the environment on health status of intensive care unit patients: Functional data analysis using wearable monitoring systems","authors":"Francisco J. Rodríguez-Cortés , Juan Antonio Arias-López , Manuel Oviedo-de la Fuente , Jose M. Jiménez-Pastor , Luna López-Coleto , Pedro Arévalo-Buitrago , Juan de la Cruz López-Carrasco , Rocío Valverde-León , Pablo Jesús López-Soto , Ignacio Morales-Cané","doi":"10.1016/j.iccn.2025.104183","DOIUrl":null,"url":null,"abstract":"<div><h3>Objectives</h3><div>Environmental factors in critical care units (ICUs) can significantly impact patient health. Traditional analysis methods often struggle with the continuous temporal data from wearable monitoring devices. This study explores the novel application of functional data analysis (FDA), a methodology well suited for continuous signals, to assess how environmental exposures influence ICU patient outcomes over time. Specifically, it examines the most impactful environmental variables. It evaluates FDA’s effectiveness in analysing continuous data from wearable monitoring devices, in contrast to traditional methods that overlook temporal patterns by using average or isolated measurements.</div></div><div><h3>Methods</h3><div>A prospective cohort study was conducted in a tertiary-level reference ICU in southern Spain, where wearable sensors were used to collect physiological and environmental data from 77 adult patients continuously. Functional data analysis (FDA) techniques were employed to explore temporal patterns in the collected signals.</div></div><div><h3>Results</h3><div>Noise intensity (decibels) was the most significant environmental variable, correlating with Richmond Agitation-Sedation Scale (RASS) scores and heart rate (HR). Functional additive models significantly improved model performance, achieving R<sup>2</sup> values up to 0.78 for RASS prediction in trauma patients. Effects varied across patient diagnosis subgroups.</div></div><div><h3>Conclusions</h3><div>FDA techniques, particularly functional additive models, better model complex relationships between environmental and physiological variables in the ICU. Environmental impacts differ across patient types, suggesting the need for specialised environmental interventions based on patient condition.</div></div><div><h3>Implications for Clinical Practice</h3><div>This study underscores the need for environmental monitoring in ICUs and highlights the potential of wearable sensors and advanced statistical analysis to optimise patient care and improve outcomes by tailoring environmental interventions to specific patient needs.</div></div>","PeriodicalId":51322,"journal":{"name":"Intensive and Critical Care Nursing","volume":"91 ","pages":"Article 104183"},"PeriodicalIF":4.7000,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Intensive and Critical Care Nursing","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0964339725002459","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"NURSING","Score":null,"Total":0}
引用次数: 0
Abstract
Objectives
Environmental factors in critical care units (ICUs) can significantly impact patient health. Traditional analysis methods often struggle with the continuous temporal data from wearable monitoring devices. This study explores the novel application of functional data analysis (FDA), a methodology well suited for continuous signals, to assess how environmental exposures influence ICU patient outcomes over time. Specifically, it examines the most impactful environmental variables. It evaluates FDA’s effectiveness in analysing continuous data from wearable monitoring devices, in contrast to traditional methods that overlook temporal patterns by using average or isolated measurements.
Methods
A prospective cohort study was conducted in a tertiary-level reference ICU in southern Spain, where wearable sensors were used to collect physiological and environmental data from 77 adult patients continuously. Functional data analysis (FDA) techniques were employed to explore temporal patterns in the collected signals.
Results
Noise intensity (decibels) was the most significant environmental variable, correlating with Richmond Agitation-Sedation Scale (RASS) scores and heart rate (HR). Functional additive models significantly improved model performance, achieving R2 values up to 0.78 for RASS prediction in trauma patients. Effects varied across patient diagnosis subgroups.
Conclusions
FDA techniques, particularly functional additive models, better model complex relationships between environmental and physiological variables in the ICU. Environmental impacts differ across patient types, suggesting the need for specialised environmental interventions based on patient condition.
Implications for Clinical Practice
This study underscores the need for environmental monitoring in ICUs and highlights the potential of wearable sensors and advanced statistical analysis to optimise patient care and improve outcomes by tailoring environmental interventions to specific patient needs.
期刊介绍:
The aims of Intensive and Critical Care Nursing are to promote excellence of care of critically ill patients by specialist nurses and their professional colleagues; to provide an international and interdisciplinary forum for the publication, dissemination and exchange of research findings, experience and ideas; to develop and enhance the knowledge, skills, attitudes and creative thinking essential to good critical care nursing practice. The journal publishes reviews, updates and feature articles in addition to original papers and significant preliminary communications. Articles may deal with any part of practice including relevant clinical, research, educational, psychological and technological aspects.